72 S.-Afr. Tydskr. Dierk. 1990,25(1) Relative population levels of aurantii on commercial citrus and adjacent bush

M.J. Gilbert * Letaba Estates, P.O. Box 6, Letaba 0870 and Department of Zoology and Entomology, Rhodes University, P.O. Box 94, Grahamstown, 6140 Republic of South

Received 15 March 1989; accepted 27 September 1989

Citrus thrips SCirtothrips aurantii Faure is a serious economic pest in southern Africa. It is indigenous, breeding on many wild hosts as well as on citrus. In this northern Transvaallowveld study, bush containing known host plants of S. aurantii was not the source of early infestation of a navel orange orchard by adult citrus thrips during September and October. Early-season damage therefore resulted from the activities of citrus thrips which had overwintered in the orchard. In contrast, S. aurantii populations on wild hosts could possibly boost orchard populations later in the season and contribute to late damage on the fruit. This is because, in one of the years, adult thrips numbers were higher in the bush than in the orchard during November and December.

Die sitrusblaaspootjie Scirtothrips aurantii is 'n ernstige ekonomiese plaag in suidelike Afrika. Dit is 'n inheemse plaag wat op baie verskillende wilde gasheerplante asook sitrus aanteel. In hierdie stud ie, wat in die Noord-Transvaalse Laeveld uitgevoer is, is aangetoon dat wilde gasheerplante geen gevaar inhou as bron van infestasie vir 'n navellemoenboord gedurende September en Oktober nie. Skade wat vroeg in die seisoen opgedoen is, is dus veroorsaak deur die aktiwiteite van sitrusblaaspootjies wat in die boord oorwinter het. In teenstelling hiermee kan S. aurantii-populasies op wilde gasheerplante moontlik boordpopulasies later in die seisoen verhoog, en dit mag bydra tot laatskade op die vrugte. Dit is omdat daar in een jaar gevind is dat blaaspootjiegetalle gedurende November en Desember hoar was in die bos as in die boord.

• Address for correspondence

Scirtothrips aurantii Faure is a polyphagous African the implications for recommended control measures and

. species recorded from over 70 host plants (Faure 1929). population monitoring procedures were also examined. )

0 It attacks several crops but is considered to be a major 1

0 pest only on citrus. Adult thrips and larvae cause Locality. Materials and Methods 2 superficial scarring by feeding on the surface of the fruit d Studies were undertaken from September to December e t rind. Severely blemished fruit are not suitable for 1986 and 1987, on Letaba Estates, a large citrus a d exporting. Damage, in the form of leaf malformation

( plantation of 1 700 ha, approximately 20 km south-east

r and stunting of growth, is also caused to young citrus e of Tzaneen in the northern Transvaal lowveld. The h s foliage. experimental site was an orchard of 200 mature i l b During an examination of the relationship between Washington navel orange trees surrounded by largely u thrips populations in citrus orchards and adjacent

P undisturbed indigenous vegetation. During the study

e natural vegetation in the western Transvaal highveld, period, the citrus orchard received no chemical treat­ h t Bedford (1943) found acacias to be amongst the most ments for the control of pests. The surrounding vegeta­ y b important indigenous hosts. Fruit on citrus trees tion contained thorn trees, (mainly the Whitethorn, d e adjacent to acacias showed more thrips damage than Acacia polyacantha Willd. subsp. campylacantha t n fruit in the centre of orchards. Although this damage was (Hochst. ex A.Rich) Brenan), interspersed with smaller a r caused during early November, it was suggested that plant species including sicklebush Dichrostachys cinerea g

e migration from natural bush into orchards might also (L.) Wight & Am. subsp. nyassana (Taub.) Brenan, and c n occur as early as August and September and thus be 'pride of the Cape' Bauhinia galpinii N.E. Br .. e c i

l responsible for the build up of S. aurantii which is r Hand sampling e commonly observed during spring. In contrast, Sam­ d n ways, Tate & Murdoch (1987) showed that densities of Thrips were collected by beating citrus trees and u this pest were consistently lower in natural vegetation adjacent vegetation to dislodge specimens onto white y a than on citrus in both the eastern Cape Province and paper, then stored in 70% alcohol. Care was taken to w e eastern Transvaal. Immigration of citrus thrips into t ensure that identifications were correct in view of the a orchards was therefore considered to be unimportant. similarity between S. aurantii and others of the same G t The aims of the present study were to determine genus (Faure 1929; Gilbert 1986). Adults were mounted e n i whether early spring damage to citrus fruit is caused by on slides for detailed examination using the method of b

a S. aurantii populations which have overwintered in the Mound & Pitkin (1972). Second instar thrips larvae were S orchards, or as a consequence of a build-up of numbers collected from plants on which S. aurantii adults had y b following migration of citrus thrips adults into the been found. In order to confirm that the larvae were S. d e orchard from adjacent bush. In the light of these results aurantii they were confined within a 'Tashiro' acrylic c u d o r p e R S. Afr. J. Zool. 1990,25(1) 73

cage (Tashiro 1967), containing host plant leaves as a slide smeared with ReverantR and supported by a larval substrate. The mature larvae pupated within the perspex or zinc plate. Thrips were counted within a cage, and the identity of the emerging adults was central area of the trap measuring 160 x 160 mm. In subsequently ascertained. order that the fluorescence of the trap did not fade as a result of direct exposure to the sun six traps were hung Trapping on the south side on the outer canopy of trees in each of Owing to the similarity in appearance between S. the two habitats. Traps were placed at a height of 2 m. aurantii and S. fulleri Faure, and the likelihood of their For commercial monitoring of thrips in citrus orchards a set ofthree traps per 5 ()()() trees is sufficient to provide a confusion during routine censuses, it was decided to reliable monitoring system (Samways 1986). Traps were monitor the occurrence of both species using attractant positioned in two rows of three, with approximately 10 m fluorescent yellow sticky traps (Samways 1986). Each between rows and a similar distance between traps trap was 25 x 25 cm and consisted of a self-adhesive within a row. The orientation of the traps was such that plastic sheet overlain with an overhead transparency those in the orchard could not be seen from the bush and vice versa. Counts of thrips were recorded weekly. Table 1 Thysanoptera collected in citrus orchard and bush by hand collection and trapping Results Thrips species recorded Citrus Bush Table 1 records thrips species collected in the citrus Species Hand Trap Hand Trap orchard and adjacent bush by trapping, and by hand Craspedothrips hargreavesi (Kamy) + + + + collection. Four Scirtothrips spp. were recorded, Haplothrips articulosus Bagnall + S. aurantii, S. fulleri, S. dorsalis Hood, and S. combreti Haplothrips bedfordi lacot-Guillarmod + + + Faure. Within the orchard the latter three species were Haplothrips clarisetis Priesner + + trapped on few occasions and not hand collected at all. Hydatothrips adolfifriederici Kamy + + S. fulleri was at times very numerous in samples taken Mycterothrips laticauda Trybom + + + from Acacia trees. Mycterothrips sp. indesc. + Thrips tenellus Trybom was common within citrus flowers along with lower numbers of Taeniothrips gow­ Scirtothrips aurantii Faure + + + + deyi (Bagnall), and T. simplex (Morison). Haplothrips Scirtothrips combreti Faure + +

.

) bedfordi lacot-Quillarmod and H. clarisetis Priesner Scirtothrips dorsalis Hood + + + 0

1 were collected on fruit, often associated with S. aurantii. Scirtothrips fulleri 0 Faure + + +

2 H. bedfordi is recorded as a citrus thrips predator Scolothrips hartwigi Priesner + + + d (Bedford 1943), but H. clarisetis is known only as a plant e

t Taeniothrips gowdeyi (Bagnall) +

a feeder (Zur Strassen 1960). Numbers of Haplothrips spp.

d Taeniothrips simplex (Morison) + (

were generally very low compared with those of S. r Thrips tene/lus Trybom + + + + e aurantii. Scolothrips hartwigi Priesner, collected on h s i l b u P

32 o ORCHARD e h t • BUSH

y 28 b d e Go t < 24 100'11. PETAL FALL n I!:

a l- r I!: g

w 20 ! Go e c n Z e =< c I!: 18 i

l ~ r

e ~I d z 12 n

2 y • a 8 w e t a 4 G t e n

i 0

b ~'-""-.-;---.---Y~'-...

a 8 16 22 29 8 13 20 27 3 10 17 24 8 Hi S

y SEPT. OCT. NOV. DEC. b

d Figure 1 Numbers of Scirtothrips aurantii adults caught on sticky traps in a citrus orchard and adjacent bush at Letaba Estates e c

u during 1986. d o r p e R 74 S.-Afr. Tydskr. Dierk. 1990,25(1)

citrus fruit, was observed to be preying on citrus red mite Comparison of citrus thrips numbers in the orchard and Panonychus citri (McGregor), and lowveld mite adjacent bush Eutetranychus anneckei Meyer. In 1986 and 1987 100% petal-fall occurred by the end of the third week in September. At this time materials Host plants of citrus thrips registered for the control of citrus thrips (e.g. tartar­ In addition to citrus, the following seven plant species emetic, triazophos, or isofenphos) would usually be were found to support S. aurantii adults and larvae in the applied in a commercial orchard, in order to prevent study area; Acacia karroo Hayne, Acacia polyacantha early thrips damage which mainly takes the form of ring­ subsp. campylacantha, Bauhinia galpinii, Caesalpinia shaped scars at the stem end of the fruit. pulcherrima (L.) Schwartz, Dichrostachys cinerea subsp. In 1986 S. aurantii numbers built up rapidly within the nyaccana, Mucuna coriacea Bak. subsp. irritans (Burtt­ orchard prior to 100% petal-fall (Figure 1). In contrast Davy) Verdc., and Ricinus communis Linnaeus. S. aurantii in the adjacent bush was first detected only

o ORCHARD 30 • BUSH •

28

22 IL

.

) 2 0 1

0 0 2 8 15 22 29 8 13 20 27 3 10 17 24 8 15 d e t SEPT. OCT. NOV. DEC. a d (

r Figure 2 Numbers of Scirtothrips aurantii adults caught on sticky traps in a citrus orchard and adjacent bush at Letaba Estates e

h during 1987. s i l b u P 26 e h

t 01988 y b

22 d e t n IL a

e II:

c W IL n e 14 c i l r e

d ~I(01 10 n Z

u

w 8 e t a G

t 2 e n i 0 b a 8 15 22 2~ 8 13 20 27 3 10 17 24 8 15 S

y SEPT. OCT. NOV. DEC. b d e c Figure 3 Numbers of Scirtothrips fulleri adults caught on sticky traps in bush at Letaba Estates during 1986 and 1987. u d o r p e R S. Afr. J. Zoo!. 1990,25(1) 75

during the third week of September and in very low compared with citrus. They may pose a potential threat numbers. Major differences in the two populations to adjacent citrus, and would account for the results of continued until the end of October. In 1987 citrus thrips Bedford (1943) at Rustenburg. The danger of high bush increased within the orchard as in 1986 (Figure 2), and populations of S. aurantii acting as sources of thrips late thrips in the bush were again detected for the first time in in the season was not evident in the results of Samways et the third week of September. However, in contrast to al. (1987). 1986, later in the season (during November and The build-up of S. aurantii on citrus earlier than in the December) S. aurantii numbers in the bush markedly bush can be related to management practices at Letaba. exceeded those in the orchard. During 1986 S. fulleri was In order to promote flowering and fruit-set of navels, common in the bush and in October considerably irrigation and fertilizer are applied during mid-July and outnumbered S. aurantii (Figures 1 & 3), but was only new vegetative growth is visible at the end of this month present in insignificant numbers throughout the year in or the beginning of August. This can immediately be 1987. used as a food source by citrus thrips which have overwintered in the orchard on out-of-season shoots and Discussion fruit. Letaba Estates is in a summer-rainfall area, and Samways, Tate & Murdoch (1986), set an economic the surrounding bush tends to be very dry with no new threshold of nine S. aurantii caught per three traps per growth during July and August. Early in the season, week; i.e. a mean of three thrips per trap. In the conditions for S. aurantii population increase are Transvaal, maintenance of catches in citrus orchards therefore more favourable on citrus than in adjacent below this figure up to the third week in November, and bush. Bearing in mind that a certain percentage of navel below 20 thrips per three traps from then on until the orange fruitlets can be at risk from thrips during the end of December, was found to result in less than 1 % of second week in September, owing to uneven blossom­ the fruit being unexportable because of thrips damage. ing, the extent of early damage will therefore be related Figures 1 & 2 show that at 100% petal-fall (during the to the size of the pest population within the orchard and third week in September), the time when thrips control the effectiveness of any applied control measures. Later normally begins, numbers of S. aurantii in the orchard flowering cultivars such as valencias will be at greater already exceeded the threshold by over 600% in 1986 risk from immigration of citrus thrips from the bush and 200% in 1987. Thus severe damage can occur to because their fruit remain smaller and more vulnerable young fruitlets very early in the season at Letaba if to scarring until later in the season. S. fulleri was found to be the species most likely to be

. chemical sprays are not timed correctly. ) confused with S. aurantii. Although this thrips was rare 0 The results show that host plants in bush adjacent to 1 within the orchard it was common on traps in the bush 0 citrus orchards do not support a large population of 2

during September and October 1986. Citrus growers d S. aurantii during September and October. The bush e wishing to monitor thrips in adjacent bush, as an aid to t population is insignificant compared with that which a pest management, would have to take care to avoid d

( develops within the citrus orchard itself. confusing these two species, especially as S. fulleri is of r

e This agrees with Sam ways et al. (1987), who found

h no economic importance. Scoring of S. fulleri as s that citrus thrips numbers were consistently higher in i l S. aurantii would erroneously increase the apparent b citrus orchards, particularly during September and

u importance of non-citrus hosts.

P October. To estimate the severity of a citrus thrips outbreak in e Bedford (1943) found that citrus fruit on trees h an orchard, counts of infested fruit beginning at petal­ t adjacent to bush was more severely damaged by thrips y fall have traditionally been made by growers. As wild b

than that on trees towards the centre of the orchard. The

d host plants are of no importance in early spring, trapping e initial thrips infestation in these orchards was low and t of citrus thrips within the orchard prior to, and during, n two dustings with sulphur further reduced numbers of a blossoming will give advance warning of the level of r

g the pest, such that no damage to fruit occurred during

infestation to be encountered before any damage to fruit e c September or October. However, during November can occur. n

e S. aurantii numbers on citrus trees bordering bush began c i l to increase and damage to fruit also occurred. This was Acknowledgements r e ascribed to thrips migrating from adjacent acacias, of

d I gratefully acknowledge Dr S.G. Compton for guidance n which the leaves were hardening off, making them in preparing this article. Dr R. rur Strassen, of the u unsuitable as a further food source for thrips. He y Senckenberg Museum Frankfurt, kindly provided many a speculated that similar citrus thrips migrations onto the identifications of thrips species. Thanks also to Prof. w e t crop might occur as early as August or September and be M.J. Samways for comments on the manuscript, and to a responsible for the initial infestation which causes early Letaba Estates for financial assistance with this work. G t season damage to fruit. The results presented here, and e n i those of Samways et al. (1987), show that this is not the References b a case. BEDFORD, E.C.G. 1943. The biology and economic S

y The results of 1987 do, however, show that, during importance of the South African citrus thrips, Scirtothrips b November and December in certain years, larger aurantii Faure. Publ. Univ. Pretoria, Series II Nat. Sci. d e

c numbers of citrus thrips may occur on nearby bush as No.7. 68 pp. u d o r p e R 76 S.-Afr. Tydskr. Dierk. 1990,25(1)

FAURE, J.C. 1929. The South African citrus thrips and five SAMWAYS, M.J., TATE, B.A. & MURDOCH, ELMA. other new species of Scirtothrips Shull. Transvaal Univ. 1986. Monitoring the citrus thrips and psylla using College, Pretoria, Bulletin No. 18. 21 pp. fluorescent yellow sticky traps - a practical guide. Citrus GILBERT, M.J. 1986. First African record of Scirtothrips and Subtropical Fruit lournal629: 9--15. dorsalis Hood (Thysanoptera: ) a potential pest SAMWAYS, MJ., TATE, B.A. & MURDOCH, ELM A. of citrus and other crops in southern Africa. 1. ent. Soc. 1987. Population levels of adult citrus thrips Scirtothrips sth. Afr. 49: 159--161. aurantii Faure (Thysanoptera: Thripidae) relative to MOUND, L.A. & PITKIN, B.R. 1972. Microscopic whole season and fruit-scarring. 1. Appl. Entomol. 104: 372-377. mounts of thrips (Thysanoptera). Entomologist's Gaz. 23: 1967. 121-125. TASHIRO, H. Self-watering cages for confining and mites on d~tached leaves. econ. Entomol. SAMWAYS, M.J. 1986. Spatial distribution of Scirtothrips 1. aurantii Faure (Thysanoptera: Thripidae) and threshold 60: 354-356. levels for one per cent damage on citrus fruit based on ZUR STRASSEN, R. 1960. Catalogue of the known species trapping with fluorescent yellow sticky traps. Bull. ent. of South African Thysanoptera. 1. ent. Soc. sth. Africa Res. 76: 649--659. 23: 321-367.

. ) 0 1 0 2 d e t a d ( r e h s i l b u P e h t y b d e t n a r g e c n e c i l r e d n u y a w e t a G t e n i b a S y b d e c u d o r p e R